J. Murray

Impact of a modified anti-thrombotic guideline on stroke in children supported with a pediatric ventricular assist device

BACKGROUND Stroke is the most feared complication associated with the Berlin Heart EXCOR pediatric ventricular assist device (VAD), the most commonly used VAD in children, and affects 1 in 3 children. We sought to determine whether a modified anti-thrombotic guideline, involving more intense platelet inhibition and less reliance on platelet function testing, is associated with a lower incidence of stroke. METHODS All children supported with the EXCOR at Stanford from 2009 to 2014 were divided into 2 cohorts based on the primary anti-thrombotic guideline used to prevent pump thrombosis: (1) the Edmonton Anti-thrombotic Guideline (EG) cohort, which included children implanted before September 2012 when dual anti-platelet therapy was used with doses titrated to Thromboelastrography/PlateletMapping (TEG/PM); and (2) the Stanford Modified Anti-thrombotic Guideline (SG) cohort, which included children implanted on or after September 2012 when triple anti-platelet therapy was used routinely and where doses were uptitrated to high, weight-based dosing targets, with low-dose steroids administered as needed for inflammation. RESULTS At baseline, the EG (N = 16) and SG (N = 11) cohorts were similar. The incidence rate of stroke in the SG cohort was 84% lower than in the EG cohort (0.8 vs 4.9 events per 1,000 days of support, p = 0.031), and 86% lower than in the previous Investigational Device Exemption trial (p = 0.006). The bleeding rate was also lower in the SG cohort (p = 0.015). Target doses of aspirin, clopidogrel and dipyridamole were higher (all p < 0.003), with less dosing variability in the SG cohort than in the EG cohort. There was no difference in adenosine diphosphate inhibition by TEG/PM, but arachidonic acid inhibition was higher in the SG cohort (median 75% vs 39%, p = 0.008). CONCLUSIONS Stroke was significantly less common in pediatric patients supported with the Berlin Heart EXCOR VAD using a triple anti-platelet regimen uptitrated to high, weight-based dosing targets as compared with the dual anti-platelet regimen titrated to PM, and without a higher risk of bleeding. Larger studies are needed to confirm these findings.

Antithrombotic therapy for ventricular assist devices in children: do we really know what to do?

The use of ventricular assist devices (VADs) in children is increasing. Stroke and device‐related thromboembolism remain the most feared complications associated with VAD therapy in children. The presence of a VAD causes dysregulation of hemostasis due to the presence of foreign materials and sheer forces intrinsic to the device resulting in hypercoagulability and potentially life‐threatening thrombosis. The use of antithrombotic therapy in adults with VADs modulates this disruption in hemostasis, decreasing the risk of thrombosis. Yet, differences in hemostasis in children (developmental hemostasis) may result in variances in dysregulation by these devices and preclude the use of adult guidelines. Consequently, pediatric device studies must include safety and efficacy estimates of device‐specific antithrombotic therapy guidelines. This review will discuss mechanisms of hemostatic dysregulation as it pertains to VADs, goals of VAD antithrombotic therapy for children and adults, and emerging antithrombotic strategies for VAD use in children.

Safety and Efficacy of Warfarin Therapy in Kawasaki Disease

Objective To describe the safety and efficacy of warfarin for patients with Kawasaki disease and giant coronary artery aneurysms (CAAs, ≥8 mm). Giant aneurysms are managed with combined anticoagulation and antiplatelet therapies, heightening risk of bleeding complications. Study design We reviewed the time in therapeutic range; percentage of international normalization ratios (INRs) in range (%); bleeding events, clotting events; INRs ≥6; INRs ≥5 and <6; and INRs <1.5. Results In 9 patients (5 male), median age 14.4 years (range 7.1–22.8 years), INR testing was prescribed weekly to monthly and was done by home monitor (n = 5) or laboratory (n = 3) or combined (1). Median length of warfarin therapy was 7.2 years (2.3–13.3 years). Goal INR was 2.0–3.0 (n = 6) or 2.5–3.5 (n = 3), based on CAA size and history of CAA thrombosis. All patients were treated with aspirin; 1 was on dual antiplatelet therapy and warfarin. The median time in therapeutic range was 59% (37%‐85%), and median percentage of INRs in range was 68% (52%‐87%). INR >6 occurred in 3 patients (4 events); INRs ≥5 <6 in 7 patients (12 events); and INR <1.5 in 5 patients (28 events). The incidence of major bleeding events and clinically relevant nonmajor bleeding events were each 4.3 per 100 patient‐years (95% CI 0.9–12.6). New asymptomatic coronary thrombosis was detected by imaging in 2 patients. Conclusions Bleeding and clotting complications are common in patients with Kawasaki disease on warfarin and aspirin, with INRs in range only two‐thirds of the time. Future studies should evaluate the use of direct oral anticoagulants in children as an alternative to warfarin.

Rehospitalization Patterns in Pediatric Outpatients with Continuous Flow VADs.

As continuous-flow ventricular assist devices (CF-VADs) are used increasingly in children and adolescents, more pediatric patients will be supported as outpatients. Herein we report the patterns of rehospitalization after CF-VAD implantation at a single center. We retrospectively reviewed the medical records of 19 consecutive patients who received CF-VADs between December 6, 2010 and November 5, 2016 and were discharged on device therapy. The frequency, duration, and indications for all hospitalizations between the time of implant hospitalization discharge and January 8, 2016 were analyzed. There were a total of 52 rehospitalization episodes in 16 (84%) patients over 5,101 (median 93, interquartile range [IQR] 38, 226) follow-up days. There were a median of two (IQR 1, 3) hospitalizations per patient. The median time to first hospitalization was 14 (IQR 7, 62) days. The most common admitting diagnoses were suspected infection 13 (28%) and suspected pump thrombosis in 8 (17%). Thirty-one (60%) hospitalizations included procedures, including seven (13%) requiring device-related surgery. Overall, 89% of postimplant discharge days were spent outside of the hospital. Children with CF-VADs can be discharged with acceptable readmission rates and significant time spent out of hospital. Most patients will be rehospitalized at least once between implant hospitalization and transplantation, often within 2 weeks of hospital discharge, with the most common indications for admission being suspected infection and suspected pump thrombosis. Device-related complications necessitating surgical intervention most frequently occur in destination therapy patients who are supported for longer periods of time.

Obesity and Premature Loss of Mobility in Two Adolescents with Becker Muscular Dystrophy After HeartMate II Implantation.

Weight gain is common after implantation of continuous-flow ventricular assist devices. Obesity can have a significant negative impact on mobility. For adolescents with Becker muscular dystrophy (BMD), for whom the ability to ambulate often persists into the mid-3rd decade, preservation of functional ability is critical. We report two cases of Thoratec HeartMate II left ventricular assist device (LVAD) implantation in adolescents with BMD for whom postoperative weight gain contributed significantly to an accelerated loss of ambulation and, in one case, driveline fracture in the context of repeated falls. As LVADs become an increasingly common therapy for endstage heart failure in adolescents with BMD, care must be focused not only on maintenance of device functionality, but also on management of aggressive weight, and preservation of ambulation, and skeletal muscle strength.

Alternative Strategy for Biventricular Assist Device in an Infant With Hypertrophic Cardiomyopathy

We report an infant with hypertrophic cardiomyopathy who underwent biventricular assist device placement with two 15-mL Berlin Heart EXCOR pediatric ventricular assist devices using an alternative atrial cannulation strategy. The systemic circulation was supported by left atrium (LA) to aorta cannulation. The LA was accessed through the right atrium by extending a 6-mm EXCOR cannula with a Gore-Tex graft connected to an atrial septal defect. The pulmonary circulation was supported with cannulation of the right atrium to pulmonary artery. This alternative cannulation strategy facilitated effective biventricular support and may be applicable to other patients with hypertrophic or restrictive physiology.

Functional status of United States children supported with a left ventricular assist device at heart transplantation

BACKGROUND As survival with pediatric left ventricular assist devices (LVADs) has improved, decisions regarding the optimal support strategy may depend more on quality of life and functional status (FS) rather than mortality alone. Limited data are available regarding the FS of children supported with LVADs. We sought to compare the FS of children supported with LVADs vs vasoactive infusions to inform decision making around support strategies. METHODS Organ Procurement and Transplant Network data were used to identify all United States children aged between 1 and 21 years at heart transplant (HT) between 2006 and 2015 for dilated cardiomyopathy and supported with an LVAD or vasoactive infusions alone at HT. FS was measured using the 10-point Karnofsky and Lansky scale. RESULTS Of 701 children who met the inclusion criteria, 430 (61%) were supported with vasoactive infusions, and 271 (39%) were supported with an LVAD at HT. Children in the LVAD group had higher median FS scores at HT than children in the vasoactive infusion group (6 vs 5, p < 0.001) but lower FS scores at listing (4 vs 6, p < 0.001). The effect persisted regardless of patient location at HT (home, hospital, intensive care) or device type. Discharge by HT occurred in 46% of children in the LVAD group compared with 26% of children in the vasoactive infusion cohort (p = 0.001). Stroke was reported at HT in 3% of children in the LVAD cohort and in 1% in the vasoactive infusion cohort (p = 0.04). CONCLUSIONS Among children with dilated cardiomyopathy undergoing HT, children supported with LVADs at HT have higher FS than children supported with vasoactive infusions at HT, regardless of device type or hospitalization status. Children supported with LVADs at HT were more likely to be discharged from the hospital but had a higher prevalence of stroke at HT.

Long-term pediatric ventricular assist device therapy: a case report of 2100+ days of support

Ventricular assist devices (VADs) have been placed as destination therapy in adults for more than 20 years, but have only recently been considered an option in a subset of pediatric patients. A 2016 report from the Pediatric Interagency Registry for Mechanical Circulatory Support revealed only eight pediatric patients implanted with a VAD as destination therapy. Herein, we report the case of an adolescent male with Becker muscular dystrophy who underwent VAD placement in 2011 as bridge to candidacy. He subsequently decided to remain as destination therapy and so far has accrued more than 2100 days on VAD support, the longest duration of pediatric VAD support reported in the literature to date.

Characteristics of deposits and pump exchange in the Berlin Heart EXCOR ventricular assist device: Experience with 67 cases

Pump exchanges are frequently required in the Berlin Heart EXCOR VAD. We intended to describe the characteristics of pump deposits in a larger patient series and evaluate if changes in our exchange procedure over time have led to increased complications. We reviewed all EXCOR pump exchanges in our institution from July 2004 to October 2014. We gathered data on size and location of pump deposits and exchange procedures. EXCOR devices were implanted in 38 children. Support was LVAD only in 22, BiVAD in 13, and SVAD in 3 cases. Sixty‐seven pumps were exchanged. The incidence of pump exchanges per month was higher for smaller pumps and for RVADs vs LVADs. Indications were visible pump deposit in 55, stroke without visible deposit in 5, incorporation of membrane oxygenator in 3, pump size change in 2, and sepsis in 1 case, respectively. Deposits were located in the outflow valve in 73%, inflow valve in 22%, pump body in 3%, and outflow cannula in 3%. EXCOR pumps are predominantly exchanged for deposits, which are most frequently located in the outflow valves. The procedure is now carried out without sedation at the bedside. No major complications were observed during exchanges.